Skip to content

Knowledge Base

Welcome to the public technical knowledge base. This repository collects evergreen technical knowledge, product and stack research, deep-dive documentation, platform comparisons, and tool catalogues.

Domains

Domain Topics Description
AI Agents OpenClaw, Hermes Agent, LLM Fundamentals, LLM Wiki, Zero Data Retention Personal AI agents, self-improving assistants, multi-channel gateways, LLM internals
APIs Web Services REST, GraphQL, gRPC, WebSocket, AsyncAPI, BFF patterns, gateways
CI/CD ArgoCD, FluxCD Pipeline definitions, GitOps, automated deployments
Databases CockroachDB, MySQL, PostgreSQL Relational engines, replication, distributed SQL
IaC OpenTofu, Pulumi, Terraform Infrastructure as Code provisioning
Infrastructure Kubernetes, Docker, OpenStack, OpenNebula, AWS, GCP, Alibaba Cloud, Tencent Cloud, Multi-Cloud Governance, AI Platform Engineering Container runtimes, VM orchestrators, public clouds, bare-metal, AI platforms
Messaging Kafka, NATS, RabbitMQ, Redpanda, Pulsar Event streaming, message queues, pub/sub systems
Networking Calico, Cilium, Flannel CNI plugins, network policies, eBPF
Observability Grafana, LGTM, VictoriaMetrics, Coroot, Monoscope, OpenObserve, SigNoz, SkyWalking, Observability 2.0 Telemetry, TSDBs, log aggregators, distributed tracing
Secrets External Secrets Operator, SOPS, HashiCorp Vault, Zitadel Secrets management and encryption
Service Mesh Envoy Gateway, Istio, Linkerd Sidecar proxies, traffic management, mTLS
Storage Ceph, Longhorn, MinIO Distributed storage, block/object/file
Tools Catalogue 578 tools across 26 categories Categorized inventory of AI/LLM, infrastructure, security, and developer tools

Observability

  • Grafana Ecosystem — The standard visualization and observability layer
  • LGTM Stack — One telemetry stack: Loki, Grafana, Tempo, Mimir, Pyroscope
  • VictoriaMetrics — Time-series database with a small footprint and high performance
  • Coroot — Automated APM with eBPF and service maps without instrumentation
  • SkyWalking — Apache APM with the eBPF Rover and a Java agent ecosystem
  • SigNoz — All-in-one observability on OpenTelemetry with a ClickHouse backend
  • OpenObserve — Logs, metrics, and traces at petabyte scale on object storage
  • Monoscope — Unified observability for self-hosted stacks
  • Observability 2.0 — Wide-event paradigm that replaces three-pillar telemetry silos

Infrastructure

  • Kubernetes — The standard container orchestration engine
  • Docker — Container runtime and multi-service development suite
  • OpenStack — Open-source cloud infrastructure stack with many options for customization
  • OpenNebula — VM-based platform for edge and infrastructure management
  • AWS — Largest public cloud — compute, storage, networking, managed services
  • GCP — Google Cloud Platform — data, ML, and Kubernetes-native services
  • Alibaba Cloud — Cloud from Alibaba with international regions similar to AWS
  • Tencent Cloud — Cloud from Tencent with gaming, video, and Southeast Asia presence
  • Multi-Cloud Governance — Policy, cost, and identity across providers
  • AI Platform Engineering — GPU scheduling, distributed compute, and inference infrastructure

Databases

  • PostgreSQL — Open-source relational database with a large feature set
  • CockroachDB — Distributed SQL for global transactional workloads
  • MySQL — Relational database that powers much of the web

Messaging

  • Apache Kafka — The standard event streaming platform
  • NATS — Multi-tenant messaging with a small footprint and edge-first design
  • RabbitMQ — Mature AMQP broker with quorum queues and streams
  • Redpanda — Kafka-API compatible C++ broker with no JVM
  • Apache Pulsar — Separated compute and storage with built-in geo-replication

IaC

  • OpenTofu — CNCF fork of Terraform with native state encryption
  • Terraform — IaC tool with the largest provider ecosystem
  • Pulumi — IaC with general-purpose programming languages

Networking

  • Cilium — eBPF-based CNI for network observability and security
  • Calico — CNI with BGP peering and network policy enforcement
  • Flannel — Overlay CNI that is simple to operate

Service Mesh

  • Istio — Mesh with Envoy sidecars and ambient mode
  • Linkerd — Rust-based mesh with a small footprint and high performance
  • Envoy Gateway — Kubernetes Gateway API implementation built on Envoy

Secrets

  • HashiCorp Vault — Secrets engine with dynamic credential leasing
  • Zitadel — Self-hosted IAM with OIDC, SAML, and per-tenant isolation
  • External Secrets Operator — Syncs external secret stores into Kubernetes
  • SOPS — Encrypts files in place with KMS, GPG, age, and other backends

CI/CD

  • ArgoCD — GitOps controller with a web UI and ApplicationSet patterns
  • FluxCD — GitOps-native continuous delivery for Kubernetes

APIs

  • Web Services — Guide to REST, GraphQL, gRPC, WebSocket, and AsyncAPI patterns

AI Agents

  • OpenClaw — Universal CLI agent with skills, MCP servers, and channel routing
  • Hermes Agent — Self-improving agent with reflection loops and the Pi runtime
  • LLM Fundamentals — Architecture, training, inference, and tuning fundamentals
  • LLM Wiki — Agent-maintained stateful markdown wikis as an alternative to stateless RAG
  • Zero Data Retention — Prevents provider-side retention of LLM interactions in regulated environments

Storage

  • Ceph — Unified distributed storage (block, object, file) at scale
  • MinIO — S3-compatible object storage with high performance
  • Longhorn — Cloud-native distributed block storage for Kubernetes

Reference Policy

  • Distilled notes over copies: We write distilled operational insights. We do not copy raw documentation. Official documentation stays the source of truth.
  • Neutral comparisons: Dedicated, neutral comparison pages track each choice. The pages contrast the architectural trade-offs.

Topic Shape

Each major technology follows the Standard Topic Folder Shape (3-5 files): - index.md — folder hub with summary, evaluation, sources, and questions (>= 80 lines) - architecture.md — component breakdown, Mermaid diagrams, how-it-works internals, and benchmarks (>= 150 lines) - operations.md — ops guide, best practices, tuning, and commands & recipes (>= 100 lines) - security.md — identity flow, threat model, access control (>= 80 lines)

How to Navigate

  • By domain: Pick a domain from the Domains table to browse all evaluated technologies
  • By comparison: Each domain has a comparisons/ folder with neutral, side-by-side evaluations
  • By tool catalogue: The Tools Catalogue provides a categorized inventory of 578 tools across 26 categories
  • Cross-cutting patterns: Topics like multi-cloud governance, landing zones, and GitOps span multiple domains. Follow the wikilinks to trace the connections.

Contributing

This knowledge base follows the Standard Topic Folder Shape with consistent conventions, placement rules, and metadata guidance across all domains.